Real-World Glycemic Outcomes of Open-Source Automated Insulin Delivery During 12 Months Among People With Type 1 Diabetes Across Age Groups.

AIMS: To assess glycemic outcomes across different age groups and explore factors associated with glycemic target achievement during 12-months open-source android artificial pancreas (AAPS) use among people with type 1 diabetes (T1D). METHODS: This study included people with T1D who had used AAPS continuously for 12 months. Glycemic outcomes were assessed at 3, 6, and 12 months post-initiation by age stratification (2-7, 8-12, 13-18, 19-30, 31-45, and > 45 years). Factors associated with achieving composite target of time in range of 70-180 mg/dL[TIR] > 70% and time below range of 70 mg/dL [TBR70] < 4% at 12 months were explored. RESULTS: A total of 190 individuals were enrolled, with median age of 26.1 (11.9, 36.7) years and T1D duration of 2.1 (0.2, 8.4) years. Mean TIR remained high and stable during 12 months (79.9% ± 11.8%, 77.8% ± 13.9%, and 77.2% ± 13.5% at 3, 6, and 12 months), with TBR70 of 3.8 (1.9, 6.1)%, 3.3 (1.6, 5.8)% and 3.6 (2.0, 6.6)% respectively. However, outcomes varied by age. At 12 months, TIR was 72.8% ± 15.5% in 13-18 years group, with TBR70 of 4.2 [2.7, 6.3]%, while > 45 years group had TIR of 78.9% ± 11.8% and relatively less hypoglycemia (TBR70, 2.5 [1.2, 5.9]%). In exploratory analyses, higher target achievement rate of the composite target was associated with longer time in automation in adults (OR 1.04; p = 0.03) and higher meal announcement frequency in paediatrics under 18 years (OR 1.35; p < 0.01). CONCLUSIONS: Open-source AAPS achieved sustained glycemic improvement across age groups in T1D people. Age-related differences in influencing factors highlight the need for age-tailored strategies to optimise long-term outcomes.